ISMRM25 - AI-Based Acquisition & Reconstruction: Part III

ISMRM & ISMRT Annual Meeting & Exhibition • 10-15 May 2025 • Honolulu, Hawai'i

AI-Based Acquisition & Reconstruction: Part III

Digital Poster

AI & Machine Learning

Tuesday, 13 May 2025

Exhibition Hall

13:30 - 14:30

Session Number: D-29

No CME/CE Credit

		Computer Number: 65 2642. Validation of Deep Learning Volume Interpolated Breath-hold Exam (VIBE) versus Standard VIBE for Detecting Internal Auditory Canal Lesions. View Presentation Video C. H. Chiang, S. Buathong, A. Hajati, A. Tabari, W-C Lo, D. Nickel, B. Clifford, R. Sellers, S. F. Cauley, J. Conklin, S. Y. Huang Department of Radiology, Massachusetts General Hospital, Boston, United States Impact: DL-VIBE supports efficient clinical workflows and reliable diagnostics compared with standard VIBE, encouraging wider adoption of DL-based MRI protocols.
		Computer Number: 66 2643. Highly-Accelerated, Free-Breathing, Time-Resolved 4D Golden-Angle Radial MRI with Self-Supervised Learning View Presentation Video H. Pei, D. Xia, Y. Wang, H. Chandarana, D. Sodickson, L. Feng New York University Grossman School of Medicine, New York, United States Impact: The proposed DeepGrasp technique allows for shorten data acquisition and efficient image reconstruction without requiring reference images for network training, providing significant potential for different clinical applications such as DCE-MRI or MRI-guided radiotherapy.
		Computer Number: 67 2644. MRI Pulse Sequence Discovery with Joint Optimization of B-spline parameterized Trajectory and Sequence Parameters View Presentation Video S. Hussain, J. Huber, M. Günther, D. Hoinkiss Fraunhofer MEVIS, Bremen, Germany Impact: This method enables design of optimized MRI sequences tuned to specific imaging objectives, opening possibilities for application-specific sampling patterns and optimized sequences across diverse imaging tasks.
		Computer Number: 68 2645. Application of AI-assisted compressive sensing technique combined with multiband technique in knee MRI scanning M. Yang, H. l. Zhang, X. Yang, Q. Zhang, Z. Lin, Y. Guo, G. Li, X. Xin Department of Radiology, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, No. 321, Zhongshan Road, Nanjing, Jiangsu, 210008, China , Nanjing, China Impact: ACS Multiband and ACS improve MRI efficiency and patient comfort while preserving diagnostic quality
		Computer Number: 69 2646. Preoperative MRI-based Deep Learning Reconstruction and Classification Model for Assessing Rectal Cancer L. Chen, Y. Yuan, S. Ren, H. Lu, F. Shen Changhai Hospital, Naval Medical University, Shanghai, China Impact: Deep learning reconstruction could improve the image quality of rectal MRI and enhance the diagnostic performance for the TN stage of rectal cancer, which could be used to promote visualization and diagnostic performance in patients with rectal cancer.
		Computer Number: 70 2647. Edge-Guided Reconstruction of Multi shot Diffusion MRI Using Projection Iteration Threshold Deep Network View Presentation Video X. Liu, H. Guo, C. Luan, H. Zhang Shenyang University of Technology, Shengyang, China Impact: This study introduced EPISTA⁺, a deep learning technique that integrates physical reconstruction model with image edge constraints to enhance the image quality of multi-shot DWI. This method improves image quality of high-resolution DWI and enhances diagnostic reliability in clinical settings.
		Computer Number: 71 2648. Deep learning-based MR image reconstruction: coil-by-coil reconstruction versus direct mapping View Presentation Video Y. Wu, M. Hu, C. Alkan, J. Oscanoa, Y. Ma, A. Syed, C. Liao, C. Huang, M. Alley, F. Zhang, A. Sun, J. Pauly, S. Vasanawala Stanford University, Stanford, United States Impact: Comparing DL-based direct mapping with coil-by-coil reconstruction and incorporating coil sensitivity in different ways are two fundamental questions in multi-coil MRI reconstruction. This work provides evidence that implicit estimation and integrated use of coil sensitivities may provide improved reconstruction.
		Computer Number: 72 2649. Deep Learning Reconstruction Enables Accelerated Prostate MRI Without Compromising Reader-Assessed Visual Quality View Presentation Video S. Fransen, Q. van Lohuizen, G. Yiasemis, J. Twilt, C. Roest, Y. Arita, J. Borstlap, J. Fütterer, M. de Rooij, D. Rouw, I. Schoots, B. Turkbey, S. Whitney, F. Simonis, J. Teuwen, H. Huisman, D. Yakar, T. Kwee University Medical Center Groningen, Groningen, Netherlands Impact: DL-based reconstruction can generate high-quality T2-weighted prostate MRIs from accelerated acquisitions without any perceived loss in overall visual quality by expert radiologists compared with the original non-accelerated images.
		Computer Number: 73 2650. Single-spiral MRE image reconstruction using zero-shot self-supervised deep learning towards real-time stiffness mapping View Presentation Video S. Martin, F. Zimmermann, J. Schattenfroh, P. Schuenke, I. Sack, C. Kolbitsch, A. Kofler Physikalisch-Technische Bundesanstalt (PTB), Braunschweig and Berlin, Germany Impact: The proposed self-supervised reconstruction ensures accurate elastogram estimation from highly undersampled data, reducing scan time. These developments promise to further enhance MRE’s impact and efficiency in clinical practice.
		Computer Number: 74 2651. Complex Swin Transformer for Accelerating Enhanced SMWI Reconstruction View Presentation Video M. Usman, S-M Gho Heuron Co., Ltd., Seoul, Korea, Republic of Impact: This research enables high-quality SMWI generation from reduced k-space data, accelerating scan times while preserving diagnostic detail. The approach could significantly enhance SMWI's clinical application for Parkinson’s Disease and support faster, more efficient neuroimaging workflows.
		Computer Number: 75 2652. Uncertainty-aware Quantitative MRI Reconstruuction using conidtional Wasserstein GAN H. Sun, Z. Li, R. Yang, Z. Xu, H. Li, X. Lin, H. Chen Tsinghua University, Beijing, China Impact: The study offers clinicians and researchers a reliable qMRI reconstruction method with pixel-wise uncertainty assessment. This could spark further investigations into model reliability and potentially facilitate the practical application of deep learning-based qMRI methods.
		Computer Number: 76 2653. Accelerated cartilage T1ρ mapping with Denoising Diffusion Probabilistic Model (DDPM) and Generative Adversarial Network (GAN) View Presentation Video R. Liu, Z. Zhang, P. Huang, J. H. Kim, M. Yang, X. Li, L. Ying University at Buffalo, Buffalo, United States Impact: We studied the combination of DDPM with GAN for accelerated T1ρ imaging where both T1ρ-weighted images and T1ρ maps are obtained simultaneously from accelerated scans. The proposed method shows improvement over the competing methods.
		Computer Number: 77 2654. Joint MRI Reconstruction and Denoising Using Noise-Adaptive Self-Supervised Learning View Presentation Video N. Janjusevic, A. Khalilian-Gourtani, Y. Wang, L. Feng Bernard and Irene Schwartz Center for Biomedical Imaging, New York University Grossman School of Medicine, New York, United States Impact: LPDSNet surpasses current methods, especially under mismatched noise-level conditions between training and testing, making it highly effective for noisy, limited-sample MRI datasets and promising for low-SNR, low-field MRI applications.
		Computer Number: 78 2655. Ultra-acceleration: compounding AI reconstruction methods with end-to-end training to achieve over fifty-fold acceleration. View Presentation Video J. Grover, S. Shan, P. Keall, D. Waddington The University of Sydney, Sydney, Australia Impact: Our work demonstrates compounding acceleration techniques can shorten the acquisition time of MRI by over 50 times. This finding is a step towards real-time volumetric MRI for interventional guidance.
		Computer Number: 79 2656. Dynamic Convolution Unrolling Networks for Cardiac Cine MR Reconstruction D. Gan, J. Cheng, Y. Zhu, D. Liang Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Impact: This study introduces an efficient tool for cardiac CINE MRI reconstruction, enhancing image quality for clinical use and supporting accurate diagnoses. It also provides valuable insights for advancing future dynamic reconstruction methodologies in medical imaging.
		Computer Number: 80 2657. Investigating the Impact of Control Information on Fidelity of Detail Recovery in Latent Diffusion Models for Undersampled MRI Reconstruction View Presentation Video X. Tang, K. Tong, L. LI, Y. ZHANG, L. YAN, J. GUAN Shenzhen Technology University, Shenzhen, China Impact: A key prerequisite for translating LDM-based MRI reconstruction methods into clinical practice is resolving the trade-off between detail richness and fidelity. Our research contributes to advancing solutions for the reconstruction fidelity challenge.

Similar Session(s)